Search NASA⌕ Search

SEARCH · Search NASA

Results for “distributed computation”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 811 records · Page 45

Flexible Pilot Jobs Framework for Distributed High Throughput Computing

Experimental particle physics has been at the forefront of analyzing the world’s largest datasets for decades. The high-energy physics (HEP) community was among the first to develop suitable software and computing tools for this purpose. GlideinWMS is a Glidein-based workload management system whose purpose is to provide experiments like CMS at CERN, DUNE at Fermilab, and others, a way to access and efficiently use vast amounts of computing resources. This system wants to provide a simple way to submit jobs to a set of computing resources, that will be provided to users behind the scenes. Glideins are the pilot jobs executed on the worker nodes at the grid sites, performing operations such as hardware detection, environment setup, and error handling. After all these operations, they will launch the actual user job. Many grid sites are supported, such as shared clusters, Google CE, and AWS. My internship aimed to design and code a flexible pilot jobs framework that will replace the one used by GlideinWMS, developing a modular and flexible skeleton of the Glidein and adding further functionalities. My project also focused on the application of machine learning techniques as support to this management system.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Computation of spanwise distribution of circulation and lift coefficient for flapped wings of arbitrary planform

The question of the effect of distribution and magnitude of spanwise circulation and shed vorticity from an airplane wing on the distribution pattern of agricultural products distributed from an airplane was studied. The first step in an analysis of this question is the determination of the actual distribution of lift along an airplane wing, from which the pattern of shed vorticity can be determined. A procedure is developed to calculate the span loading for flapped and unflapped wings of arbitrary aspect ratio and taper ratio. The procedure was programmed on a small programmable calculator, the Hewlett Packard HP-97, and also was programmed in BASIC language. They could be used to explore the variations in span loading that can be secured by variable flap deflections or the effect of flying at varying air speeds at different airplane gross weights. Either an absolute evaluation of span loading can be secured or comparative span loading can be evaluated to determine their effect on swath width and swath distribution pattern. The programs are intended to assist the user in evaluating the effect of a given spanload distribution.

Razak, K.↗

Computational-experimental pressure distributions on a transonic, low-aspect-ratio wing

A generic, transonic, supercritical, low aspect-ratio wing was tested at design incidence of 5 deg at Mach numbers from 0.25 to 0.96. Oil-flow studies at the design Mach number of 0.85 showed local-flow separation, which, in retrospect, might have been avoided. At Mach 0.82 with unseparated flow, the surface-flow angles were less than 10 deg. Predictions with the FLO22 transonic potential code are good. Lift interference is strong without tunnel-wall suction. Evidence from this study shows that wings that are optimized for mild shock waves and pressure-recovery gradients generally have small, three-dimensional flow at conditions for unseparated flow.

Keener, E. R.↗

Human-computer interaction in distributed supervisory control tasks

An overview of activities concerned with the development and applications of the Operator Function Model (OFM) is presented. The OFM is a mathematical tool to represent operator interaction with predominantly automated space ground control systems. The design and assessment of an intelligent operator aid (OFMspert and Ally) is particularly discussed. The application of OFM to represent the task knowledge in the design of intelligent tutoring systems, designated OFMTutor and ITSSO (Intelligent Tutoring System for Satellite Operators), is also described. Viewgraphs from symposia presentations are compiled along with papers addressing the intent inferencing capabilities of OFMspert, the OFMTutor system, and an overview of intelligent tutoring systems and the implications for complex dynamic systems.

Mitchell, Christine M.↗

Measurements over distributed high performance computing and storage systems

A strawman proposal is given for a framework for presenting a common set of metrics for supercomputers, workstations, file servers, mass storage systems, and the networks that interconnect them. Production control and database systems are also included. Though other applications and third part software systems are not addressed, it is important to measure them as well.

Williams, Elizabeth↗

The alignment-distribution graph

Implementing a data-parallel language such as Fortran 90 on a distributed-memory parallel computer requires distributing aggregate data objects (such as arrays) among the memory modules attached to the processors. The mapping of objects to the machine determines the amount of residual communication needed to bring operands of parallel operations into alignment with each other. We present a program representation called the alignment distribution graph that makes these communication requirements explicit. We describe the details of the representation, show how to model communication cost in this framework, and outline several algorithms for determining object mappings that approximately minimize residual communication.

Chatterjee, Siddhartha↗

The alignment-distribution graph

Implementing a data-parallel language such as Fortran 90 on a distributed-memory parallel computer requires distributing aggregate data objects (such as arrays) among the memory modules attached to the processors. The mapping of objects to the machine determines the amount of residual communication needed to bring operands of parallel operations into alignment with each other. We present a program representation called the alignment-distribution graph that makes these communication requirements explicit. We describe the details of the representation, show how to model communication cost in this framework, and outline several algorithms for determining object mappings that approximately minimize residual communication.

Chatterjee, Siddhartha↗

Application of a distributed network in computational fluid dynamic simulations

A general-purpose 3-D, incompressible Navier-Stokes algorithm is implemented on a network of concurrently operating workstations using parallel virtual machine (PVM) and compared with its performance on a CRAY Y-MP and on an Intel iPSC/860. The problem is relatively computationally intensive, and has a communication structure based primarily on nearest-neighbor communication, making it ideally suited to message passing. Such problems are frequently encountered in computational fluid dynamics (CDF), and their solution is increasingly in demand. The communication structure is explicitly coded in the implementation to fully exploit the regularity in message passing in order to produce a near-optimal solution. Results are presented for various grid sizes using up to eight processors.

Deshpande, Manish↗

Effects of Spatial Gradients on Electron Runaway Acceleration

The runaway process is known to accelerate electrons in many laboratory plasmas and has been suggested as an acceleration mechanism in some astrophysical plasmas, including solar flares. Current calculations of the electron velocity distributions resulting from the runaway process are greatly restricted because they impose spatial homogeneity on the distribution. We have computed runaway distributions which include consistent development of spatial gradients in the energetic tail. Our solution for the electron velocity distribution is presented as a function of distance along a finite length acceleration region, and is compared with the equivalent distribution for the infinitely long homogenous system (i.e., no spatial gradients), as considered in the existing literature. All these results are for the weak field regime. We also discuss the severe restrictiveness of this weak field assumption.

MacNeice, Peter↗

A Comparative Study on Computation of Cumulative Distribution Function in Predicting Time of Failure of Engineering Systems

Estimating accurate Time-of-Failure (ToF) of a system is key in making the decisions that impact operational safety and optimize cost. In this context, it is interesting to note that different approaches have been explored to tackle the problem of estimating ToF. The difference is in part characterized by different definitions of the hazard zones. The conventional definition for the cumulative distribution function (CDF) calculation is assumed to have well-defined hazard zones, that is, hazard zones defined as a function of the system state trajectory. An alternate method suggests the use of hazard zones defined as a function of the system state at time , instead of hazard zones defined as a function of system state up to and including time k (Acuna and Orchard 2018, 2017). This paper explores these differences and their impact on ToF estimation. Results for the conventional CDF definition indicated that, (i) the cumulative distribution function is always an increasing function of time, even when realizations of the degradation process are not monotonic, (ii) the sum of all probabilities is always 1 and does not need to be normalized, and (iii) all probabilities are positive and less than or equal to 1. Similar results are not observed for CDF calculation with hazard zones defined as a function only of the system state at time . Results for ToF estimation using Acuna's definition differ, suggesting that there is an underlying assumption of independence in the hazard zone definition. Therefore, we present an alternate definition of hazard zone which guarantees the properties of a well-defined CDF with a more straightforward ToF definition.

Hazard Zone↗

Derivation of Integrated Load Distributions from Resampled Computational Data

Principal component analysis (PCA) has been the center of many surrogate models used to characterize fluid flows in recent years. However, little work has been done to character- ize the uncertainty in the PCA transform itself and its effect on derived surrogate models. To explore the uncertainty, a typical interpolated surrogate model is constructed for a represen- tative aerodynamic body from computational data. The computational data is then resampled to explore the robustness of the PCA transformation. The variations of the model predictions during this resampling are analyzed to get a measure of confidence in the PCA transformation, which is then applied to the interpolated surrogate model to get uncertainty on integrated force predictions. An initial test case has been explored with promising results.

Principal Component Analysis↗

Derivation of Integrated Load Distributions from Resampled Computational Data

Principal component analysis (PCA) has been the center of many surrogate models used to characterize fluid flows in recent years. However, little work has been done to character- ize the uncertainty in the PCA transform itself and its effect on derived surrogate models. To explore the uncertainty, a typical interpolated surrogate model is constructed for a represen- tative aerodynamic body from computational data. The computational data is then resampled to explore the robustness of the PCA transformation. The variations of the model predictions during this resampling are analyzed to get a measure of confidence in the PCA transformation, which is then applied to the interpolated surrogate model to get uncertainty on integrated force predictions. An initial test case has been explored with promising results.

Computational Fluid Dynamics↗

CUDA Computation of the Feynman Distribution

In 2006, NVIDIA Corporation introduced the Computer Unified Device Architecture (CUDA) for C, C++, and Fortran programming languages. The CUDA parallel computing platform uses the Graphics Processing Unit (GPU) of NVIDIA graphics card, and it is designed for parallel computations. In these computations, the same algorithms are executed on the data in parallel. The GPU contains many Streaming Multiprocessors (SM) that can independently execute blocks of threads. A thread is a sequence of serial programmed instructions. The SM executes threads in groups of 32 and a group of 32 threads is called warp. The threads within a block can be synchronized and have access to a shared memory that is visible only to them. Conversely, blocks of threads cannot be synchronized and have access only to the global memory of the graphics card. This memory is called device memory and is independent of the host memory controlled by the Central Processing Unit (CPU). A single thread has access to a local memory (register) that is not visible to any other thread. The part of the program that is executed on the graphics card is called kernel. Each thread of the kernel has a unique identification index that is accessible by the in-built variable threadIdx; the latter is a 3-component vector. A threads block can contain up to 1024 threads. Blocks of threads have a unique identification index that is accessible by the in-built variable blockIdx; the latter is a 3-component vector. The in-built variable blockDim contains the number of threads per block; this variable is a 3-component vector. Threads and blocks represent different levels of parallelism; a kernel is executed as a grid of blocks of threads. Figure 1 illustrates a grid of thread blocks for a GPU with 2 SM. A SM can execute multiple blocks of threads in parallel; a block of threads is executed on a single SM. A GPU can execute multiple kernels concurrently. This work has used the Quadro M5000 graphics card with the following technical specifications: - 2048 cores with 861 MHz frequency, - 16 streaming multiprocessors, - 5.2 trillion transistors, - 8.2 Gb device memory, - 64 kb register memory (per thread), - 48 kb shared memory (per block) - 211.6 GB/s memory bandwidth, and - 3.5 GFLOPS floating point performance.

97 MATHEMATICS AND COMPUTING↗